NSE Quant Engine

A desktop screener and analytics engine for NSE (Indian equities & ETFs), plus a minimal TanStack Start web shell used only so the Python source can live in a normal Lovable / GitHub repository.

Honest note: this is a screener that, once validated, can tell you whether its ranking beats a benchmark after costs. It is not a low-risk/ high-profit/short-hold oracle. The validation layer's most valuable possible answer is sometimes "No Proven Edge Yet." Believe it when it says so.

Repository layout

desktop/nse_quant_engine/   # The actual app — Python + PySide6 desktop UI + engine
src/                        # TanStack Start web shell (placeholder preview only)

The web app is intentionally minimal. All analytical logic lives under desktop/nse_quant_engine/.

Quick start — desktop engine

Requires Python 3.11+.

cd desktop/nse_quant_engine
pip install -r requirements.txt
python run_app.py

Windows: double-click run_app.bat. macOS: run_app.command. Full pipeline (headless): run_full_workflow.bat or python orchestrator.py. Shadow-mode (dormant adaptive layer): run_shadow_mode.bat.

Runtime folders data/ and output/ are created on first run and are not committed.

Tests

cd desktop/nse_quant_engine
python -m pytest tests/ -q

Documentation

Deeper docs live next to the code:

• desktop/nse_quant_engine/README.md — engine overview & design
• desktop/nse_quant_engine/QUICKSTART_WINDOWS.md — first-run walkthrough
• desktop/nse_quant_engine/WORKFLOW.md — the 16-step pipeline
• desktop/nse_quant_engine/INTEGRATION_GUIDE.md — how modules bolt together
• desktop/nse_quant_engine/INSPIRATION_MAP.md — concept → module → artifact table

Web shell (optional)

bun install
bun run dev        # http://localhost:8080
bun run build

Credits

Professional-desk features (macro regime, sector/peer context, event calendar, FII/DII + bulk-deals institutional flow, multi-alpha IC survivorship, walk-forward backtest, EV/Kelly, portfolio ship-gate, regime-conditional alpha tilt, turnover-vs-cost rebalance diff, portable LLM evidence bundle) were inspired by:

• Fincept Terminal
• Vibe Trading

No code from those projects is bundled — only concepts. Terminal log lines for borrowed steps are prefixed [fincept] or [vibe] so each run makes the provenance visible.

License

MIT — see LICENSE.

Disclaimer

Provided for research and educational use. Not investment advice. Markets carry risk; past performance does not guarantee future results. You are solely responsible for any decisions made using this software.